Search

Edible insects are characterised by high microbial numbers of which the bacterial endospores fraction can survive (thermal) processing. It is unknown, however, which bacterial species occur as endospore in edible insects and what impact they have on food safety. Additionally, edible insects have never been explored for the presence of foodborne viruses so far. In this study, we found that the bacterial endospore fraction in a collection of mealworm and cricket samples obtained from commercial insect producers can comprise a large amount of Bacillus cereus group members that can pose insect or human health risks. Monitoring and effective mitigation of these risks are urged. By contrast, none of the foodborne viruses hepatitis A virus, hepatitis E virus and norovirus genogroup II were detected in the sample collection. Therefore, food safety risks originating from these viral pathogens can be concluded to be low.

The aim of this study was to determine the effect of fermentation of mealworms (Tenebrio molitor) with commercial meat starters cultures on the functional properties of powders produced from the larvae. Full fat and defatted powder samples were prepared from non-fermented and fermented mealworm pastes. Then the crude protein, crude fat and dry matter contents, pH, bulk density, colour, water and oil binding capacity, foaming capacity and stability, emulsion capacity and stability, protein solubility, quantity of free amino groups and protein composition of the powders were evaluated. Regardless of the starter culture used, fermentation significantly (p < 0.05) reduced the crude and soluble protein content of the non-defatted mealworm powders and in general impaired their water and oil binding, foaming- and emulsifying properties. Defatting of the powders improved most functional properties studied, except the protein solubility, water binding capacity, foaming capacity and emulsion stability. The o-phthaldialdehyd assay revealed that the amount of free amino groups increased during fermentation, which may be attributed to proteolysis of mealworm proteins by the starters. Sodium dodecyl sulfate polyacrylamide gel electrophoresis demonstrated that the soluble proteins of fermented powders were composed of molecules of lower molecular mass compared to non-fermented powders. As the molecular sizes of the soluble proteins decreased, it is clear that also the protein structure was modified by the fermentation process, which in turn led to changes in functional properties. It was concluded that fermentation of mealworms in general does not contribute to the functional properties studied in this work. Nevertheless, the results confirmed that the properties of non-fermented powders are comparable to other food protein sources.

The potential of insects as novel protein source in food and feed is gaining increased attention in western countries. Insects are considered as "minilivestock" with potential of being produced more sustainably as compared to traditional livestock and livestock feed ingredients (e.g. soybean meal and fishmeal), while exhibiting equal nutritional qualities. However, little knowledge exists on the endogenous microbiota of industrially reared insects and its changes during the rearing process. Furthermore, little is known about potential hazards that may affect the microbiological safety of harvested insects. This dissertation aimed to characterise the microbiota of insects during rearing and during post-harvest procedures at laboratory, large and/or industrial scale. Potential microbiological safety risks were identified by studying a selection of food pathogens during rearing. In a first series of studies, rearing cycles of three different insect species were characterised for their endogenous microbiota and its dynamics. For black soldier fly (BSF) larvae (Hermetia illucens), reared for their potential use in animal feed, rearing cycles were monitored at four different locations: one at laboratory scale and three at large scale. Lesser mealworms (Alphitobius diaperinus) and tropical house crickets (Gryllodes sigillatus) were investigated during rearing at industrial scale for human consumption. To this end, samples were taken for all three species of the insects themselves, as well as of the rearing substrates prior to administration, and of the residues in the rearing containers/cages (i.e. leftover substrate, faeces and exuviae). Intrinsic parameters, microbial quality (through plate counts) as well as bacterial community composition (through high-throughput 16S rRNA gene sequencing) were assessed. For all three insect species, a large portion of bacterial species observed in the insects was also present (although often in very different abundances) in the substrates. It appeared that the substrate was an important source of bacteria for the insect microbiota. Nevertheless, both microbial numbers and bacterial community compositions differed to large extent between substrates and insects, especially for BSF larvae and lesser mealworms. In addition, even for BSF larvae reared with different substrates and at different facilities, large differences were observed in their microbiota. Thus, the insect bacterial community composition was not merely a reflection of the microbiota in the substrate, but was likely the result of a selective process determining the ability of specific bacterial species to colonize the insect gut. As was shown for lesser mealworms, the establishment of a stable bacterial community may occur only after some weeks during the rearing process. Still, for all species studied, bacterial genera were recovered in our study that were also reported in other rearing cycles (for BSF) or in studies by other authors. Although more research is certainly necessary, the preference for certain bacterial genera or species, possibly even exhibiting functional roles in the insect gut, could be hypothesised. Samples of BSF larvae, lesser mealworms and tropical house crickets at external facilities at large and/or industrial scale were assessed for the presence of a selection of four food pathogens. Listeria monocytogenes and coagulase-positive staphylococci were never detected. Presumptive Bacillus cereus and Salmonella sp. were not detected in lesser mealworm and tropical house cricket rearing, but were detected in larvae and residue samples of BSF rearing. More specifically, Salmonella sp. was detected (present in 25 g) in the residue of one rearing facility, while presumptive B. cereus was detected in one residue sample of a second (200 cfu/g) and both residues and larvae of a third rearing facility (up to 3.8 log cfu/g). Thus, specific attention should be paid to these pathogens when BSF larvae are to be used as feed ingredients. Monitoring microbial contamination in the substrates may play an important role in assuring the absence of food pathogens in reared insects. With regard to the latter hypothesis, the potential that food pathogens possibly present in the substrate are taken up by the insects (horizontal transmission), was studied. This was assessed in a case study on the transmission potential of Salmonella sp. present in wheat bran as a substrate for yellow mealworms (Tenebrio molitor). To this end, Salmonella sp. was artificially inoculated into wheat bran in laboratory scale rearing trays at different contamination levels and its presence in the bran and larvae was determined during seven days. Results showed Salmonella sp. to remain viable in the bran for seven days in the absence of larvae, but its number was reduced in the presence of larvae. Larvae did become contaminated with Salmonella sp., but also here, its number was reduced by day 7. For the lowest inoculation level (2 log cfu/g), no Salmonellae were detected after seven days in the larvae. Thus, it appears that the risk related to the presence of Salmonella sp. in mealworms may depend on the contamination level in the bran. Nevertheless, the study shows that mealworms can become contaminated with Salmonella sp. when it is present in the bran. Monitoring of the pathogen in the bran after arrival at the rearing facility and in harvested mealworms is thus advised. After harvest, a variety of treatments may be applied to insects, such as rinsing, starvation (to empty the gut), and heat treatments. For the yellow mealworm, the impact of starvation and rinsing was investigated in detail. It was shown that neither procedure, nor a combination of both procedures, enhanced the microbial quality by reducing microbial numbers. In addition, starvation did not substantially alter the bacterial community composition. Thus, these procedures appear redundant from a microbiological point of view. After the rearing of both lesser mealworms and tropical house crickets for human consumption, a post-harvest heat treatment was applied. These treatments were shown to reduce most microbial numbers. Although the effect of heat treatments was not assessed for BSF larvae, also here, a heat treatment or other decontamination technology prior to or during further processing is advised in order to eliminate potentially present food pathogens. However, bacterial endospores were hardly affected by the treatments applied to lesser mealworms and tropical house crickets. Bacterial spores seem present in insects at high numbers. In this dissertation, numbers of up to 7.5 log cfu/g were observed. Consequently, rearers are advised to validate heat treatments with respect to spore inactivation, even though no legislative criterion exists for endospores. In particular because presumptive B. cereus was encountered in insects in this study and in insects in other studies, the risk for the presence of (spores from) this pathogen should be determined. This PhD dissertation provides general insights into the microbiota of a selection of industrially reared insect species. The results suggest that monitoring of the microbial contamination of the substrate, as well as a validation of spore-reducing post-harvest procedures, are important points of attention for the development of good hygiene practices for the developing insect sector. Based on this dissertation, future research should focus (1) on exploring the microbiological safety of (currently not authorised) substrates, probably allowing more organic waste streams to be authorised, (2) on exploring the variability of the microbiota during subsequent rearing cycles under identical and varying conditions, (3) on characterising the mycoflora and presence of mycotoxins in industrially reared insects, (4) on assessing the influence of factors so far not investigated, such as the "house flora" in an insect rearing facility, (5) on transmission of other food pathogens, such as B. cereus, to mealworms as well as to other insect species in order to provide valuable risk assessments for specific insect-pathogen combinations, (6) on developing and validating techniques to reduce (in particular) the number of endospores in insects as well as preventing germination of spores and growth of the vegetative cells in end products, and (7) on studying decontamination technologies for the residue towards allowing new applications. status: published